Abstract

An algorithm for assessing forest fire potential is tested for Kalimantan Island, Indonesia. It is based on a fuel model map modified from the US-National Fire Danger Rating System (US-NFDRS), Normalized Difference Vegetation Index (NDVI), and weather data. The Indonesian fuel model map was derived using the global 4-minute land cover data set consisting of 13 classes. The NDVI data were derived from the global 4-minute NOAA-AVHRR data. The output is presented as a monthly Fire Potential Index (FPI) from 1981 to 1993 and compared with trends in fire occurrences over the same time period. A case study illustrates correlation between the FPI and the hot-spot distribution derived from AVHRR data, as well as between the FPI and the Total Ozone Mapping Spectrometer (TOMS) Aerosol Index.